I think making hundreds of billions quickly is quite challenging even with perfect knowledge. Doing it quickly by any kind of active trading is going to be virtually impossible, because somebody needs to be losing all that money to you. The entire daily volume of the U.S. stock market is around $100 billion notional, so the maximum you could possibly make each day if you were on the right side of every single trade is a few percent of that. You can’t just say “use options”, because again the sheer size of the p&l you need to generate is the constraint - if you’re going to make a billion dollars in a day on options, somebody needs to lose that money to you.
So I think to make hundreds of billions, you just need to buy the best stocks, use margin and options for leverage, and hold the long stock positions for longer periods, say months at a time. That’s not a zero sum game, nobody is losing money to you if you buy assets and they go up.
This is going to be a lot quicker in a bull market, but obviously your perfect stock-picking capability makes it much easier.
Another point is that, even if you manage to avoid all legal trouble (and extra-legal trouble, like business rivals trying to get you assassinated), there will still be consequences. A market that includes someone with perfect information will not, in the long term, behave in the same way as one without. We don’t live in a world with a perfect trader, so we don’t know what your market would come to look like. In fact, I’m pretty sure that we can’t even know what such a market would look like.
Exactly. Or alternatively, we have to assume that you accumulate wealth slowly enough that the impact on the market is not significant. In which case, it takes a very long time.
Are companies allowed to ban someone from trading just because he’s too good? i.e., could Boeing say, “We refuse to let (hypothetical perfect-knowledge trader) buy or sell any shares of Boeing stock from this point on?”
It’s not a question of a “ban”. If you want to trade, somebody must be willing to take the other side of your trade. If you are large enough to be noticed, and everyone knows you have these magical powers, who will be willing to trade with you for a guaranteed loss?
As Chronos says, this is a situation where the market simply could not function the way it does in the real non-magical world.
It’s like one of the faster-than-light travel physics threads. You can’t just say - what if I could go faster than light, but everything else in physics was the same, then what would happen? It doesn’t work that way - if you “break” something that fundamental, then everything changes.
I suspect they still have regulatory approval hurdles that would slow you down more than owning the bank would speed things up.
It’s hard to know exactly what the OP was looking for but trying to answer your question, which is basically how quickly can this be done using the best possible trade every day, it still winds up looking a lot my hypothetical. The best trade for a minusucle trader is penny stocks on Robinhood, where you could make multiples of your money every day. Later, it will be bigger more liquid stocks where a single block trade on the tape might be 10,000 shares that moves up a few bucks over the course of a day. Even leveraged, this won’t necessarily be multiples of your money, but it will probably be greater than 10%, until you have co much money that there are no single trades that can suck up your capital. Options will eventually get you more volatility to exploit, but if you are only doing a single trade, you need to pick a contract that is both volatile and sufficiently high volume that whatever price occurs is enough to move the needle. When you ultimately have billions of dollars to commit on a single trade, the only way to do it as a practical matter is with swaps.
On these days, you short stocks or buy puts and sell calls. If you have perfect information, you make your money on volatility, regardless of the direction.
This is absolutely true.
My hunch, less than a year but more than a couple of months. Two years should definitely do it.
You can do this with futures contracts. But the OP said perfect information about the stock market so I didn’t even touch on this.
You could make way more than double your money in the early days. In the later days, as with the stock market, it becomes harder and harder to find bets big enough to make the same returns. There is no way if takes 45 years with perfect information to become the richest person in the World. Jeff Bezos and Elon Musk both did it in the real world with no superpowers before they were 50. They didn’t start when they were five.
I don’t have the number handy and I’m not sure anyone tracks it regularly but I think NYSE volume alone far exceeds $100 billion. Trading volume is around 5 billion shares per day and I don’t know the average share price but I’m guessing its closer to $100 per share than $20.
Your point is well taken that you can’t capture all of that movement though, especially if you are trading manually.
Agreed.
They absolutely cannot in the public equity markets. I’m also not sure what would motivate them to do so, even under our obtuse hypothetical. The companies themselves aren’t harmed (or even affected in any meaningful way) by the OP’s trading.
Damn, I forgot to leave out the tip jar! Thanks. I’m just glad people were interested. It was a long post and I figured for most it would be TLDR.
Sure, Bill Gates did it too but they did it though owning and running a single company. Let’s say you had perfect knowledge to invest in Amazon at its IPO in 1997 @18/share you would have seen massive growth over the last 24 years so the theoretical 100mm I posited the OP would have when he started getting market rate returns he would have $120B today. That obviously assumes a by and hold strategy not perfectly timing the peaks and dips.
The other part of that is the with their $100mm the OP would have bought 1/3 of Amazon at IPO which is likely not possible since Bezos owned only 42%. Further you’re assuming there is another Amazon coming to market in the next month. That is possible but it seems about as unlikely as the super powers the OP is talking about getting.
20% annual return after fees and taxes is phenomenal and while it may not be the peak of perfection it’s going to be pretty close.
Quoting from the book “The Man Who Solved The Market - How Jim Simons launched the Quant Revolution”, by Gregory Zuckerman, Annex 1, page 331 f.) the returns of the Medallion Fund before fees:
1988 16.3%
1989 1.0%
1990 77.8%
1991 54.3%
1992 47.0%
1993 53.9%
1994 93.4%
1995 52.9%
1996 44.4%
1997 31.5%
1998 57.1%
1999 35.6%
2000 128.1%
2001 56.6%
2002 51.1%
2003 44.1%
2004 49.5%
2005 57.7%
2006 81.4%
2007 136.6%
2008 152.1%
2009 74.6%
2010 57.5%
2011 71.1%
2012 56.8%
2013 88.8%
2014 75.0%
2015 69.3%
2016 68.6%
2017 85.4%
2018 76.4%
That is an average return 66.1% before fees, and as the main investors in the fund were the owners of the fund themselves, the (exorbitant! 5% yearly management fees, 20% performance fees 1998 - 2000, 35% performance fee 2001, 44% performance fees 2002 - 2018) fees went into they own pockets and were reinvested. So there seems to be someone that beat your lousy 20% p.a. every single year between 1990 and 2018, even during the dot.com meltdown and during the subprime crisis. Actually, they performed best 2000 and during the subprime crisis.
Pity some of the fund owners are despicable people and used the money to corrupt society. Bob Mercer, for instance.
I am afraid there is only one way to find out: actually do it. So, if someone is so kind to provide me with this perfect foresight, I will take the plunge and try it out. For the team! I promise I will use use the money more wisely than Mr. Mercer. And everybody will own a share! I’ll just skip the boring part before you reach the first million, that is a purely academic exercise with no relevance to the problem we are discussing now: it is clearly feasible in under three months. From the first million on the trouble with regulators and autorities can come hard and fast, that I am willing to find out what the best way to avoid that trouble is.
Considering that they did it for 28 years and got up to 44% of other people’s returns, returns that were worth billions, I reckon the fund managers will one day be by far the richest people in their cementery.
So the fastest way the OP is looking for may be to manage other people’s* money very well and take an exorbitant cut.
But that is outside the bounds of the OPs hypothetical because you are starting and running a business. There are lots of business that could vault you to $200+B if you run them successfully but we’re not longer talking about:
It would only be a side hustle! I mean: fair enough. Only that nobody trades manually, they use computers. And computers can be programed, we would have to discuss from what speed onwards this would be considered High Frequency Trading, and so on and so forth. But you are correct, I am going beyond the OP, so I’ll stop.
You know, that makes me wonder: let’s say that I’m rich, and that you have this 100% knack for knowing which stocks will go up and down; and let’s say that you bet me a dollar about a stock, and that I of course lose to you.
…and let’s say you do it again, and do it again. Sure, I’ll keep losing a dollar to you each time; but, upon placing that bet with you, can I then invest the opposite way? And can you start raising your rates, taking $5 or $10 off me each time I gladly lose to you while investing the way you bet?
It depends on what being “all knowing” wrt markets means. If I know how much you are investing on my tips, and I know everything about the markets so I might well know what people buy, sell, and how much, then at least the Medallion folks felt able to ask for a 44% cut + 5% management fee and got away with it and people begged them to take their money.
Did you ever read about the scam that starts sending one million people an e-mail, half of them claiming that a certain stock (or index, or fund, or whatever) would raise the next day, the other half claiming the opposite? You were right 500.000 times, to those you write again (for free!), to half of them that such and such will rise, to the other half that it will fall. Repeat 20 times, and you are down to one person to whom you have correctly predicted 20 times in a row what the market would do. What can you get this person to do?
Great and informative post, Tired_and_Cranky. You covered all the scenarios that I could think of, as well as a lot more that I hadn’t considered. It’s like a more comprehensive and better version of Randall Munroe’s What If? articles. I almost feel bad that you’re not charging people for access to these kind of posts.
Regarding what you said about options having more leverage than equities, I recall back in January 23, 2013, the day before their quarterly earnings report, Netflix’s stock price was ~$98, and analysts were pretty negative on their long-term prospects. The $125 strike price call option that expired on January 25, 2013 costed only $2 that day because it was way out of the money, and the time value of that particular option was practically nil.
However, Netflix delivered an earnings report the next day (January 24) that showed a surprise profit and rosy predictions for the future, shocking analysts and short-sellers. The stock price closed 42.2% higher in just one day, and on January 25 it reached an intraday high of ~$172, meaning those $125 call options were now deeply in the money, and would have been trading around $47 at that intraday high. Meaning if you bought 500 of those options for a total of $1,000, you could’ve made (500 options * $47 per option * 100 shares controlled per option contract) = $2.35 million, a factor of 2,350X !!!
However, like you said, even the option markets for highly-traded companies eventually reach a liquidity limit. Buying 500 of those Netflix options might have been doable, but if you’re trying to become a billionaire by buying a million of them? Forget about it.
There have been infrequent cases where a stock price closes at the end of the day, unchanged from the previous day. It’s like the 0 and 00 on roulette wheels that gives the house a built-in advantage. Another possibility is that the stock price changes so slightly from the previous day, that you’d lose more in trading commissions then you get from buying or shorting the stock.
But even Jeff Bezos didn’t have perfect market information. He couldn’t buy on the dips and sell at the short-term peaks, over and over again, to maximize his return on that stock. A 20% return, in the real world, is phenomenal but not unachievable. Several people, in fact, have done it. None of them had perfect information.
Why would he waste his time betting a dollar when he can bet millions? Also, he would figure out your game probably before the first round, and he could just fuck with you by letting you win half the time. Preventing the information leakage is worth way more than the dollar.
Turns out, I don’t have good access to historical option price data working from home but I used a Black Scholes pricing calculator and some “highest returning stocks” lists for a couple of quarters to get a lowball estimate of how much you could make in a quarter with options. With that information, buying out of the money calls, you almost couldn’t make less than a 500% return in a quarter. To boost returns even further, you sell puts on big gainers to fund your entire portfolio. You would have a license to print money.
For what it’s worth and because I am great at procrastinating, I tried to figure out roughly the maximum amount of money you could make in 6 months on a single swaps contract, trying to balance a need for low starting capital and maximizing total return.
Short version - I think you could have made $200 billion in the last six months alone, starting with something in the neighborhood of, I think, $10 billion.
I assumed that a total return swap would be the best vehicle. In a total return swap, you promise you will pay your counterparty all of the money that they would have made on some hypothetical pool of assets. In turn, they agree to pay you the hypothetical return on some other pool of assets. Those pools of assets can be basically any financial assets. I think on this board, I have joked about things like a Zimbabwean bond/Pokemon card total return swap. This is absurd but everyday, firms enter total return swaps where they swap returns on one portfolio of stocks for another portfolio of stocks. You see where this was going.
So my first thought was, I will swap a portfolio of the best performing stocks for the worst performing stocks. Seems obvious, huh? I started with lists of the best performing YTD stocks (Gamestop, Tecnoglass, Timkensteel, Veritiv, etc.) and worst performing (Sherwin Williams, Clearwater, First Solar, etc.). It turns out the obvious answer is wrong but we’ll get to that in a moment.
In a total return swap (or really any swap) the parties agree to continuously exchange money back and forth as one side makes money on his side of the deal and the other loses it. That means, I don’t just pay for a contract up front and collect my winnings later. There is a continual process of marking-to-market as the asset prices move up and down and settling the obligations. My swap dealer will want me to post enough capital to satisfy the losses I might experience if my portfolio goes catawampus. If my posted capital gets too low, the dealer will demand even more capital. They will also charge a bunch of money up front to cover another of their costs - the cost to hedge the risk I’m pushing off on them.
My goal is to get the cheapest swap I can get that will earn me $200 billion, both in terms of the upfront charge and the capital required. My swap will be cheapest if it looks safe to the dealer and if it doesn’t look expensive to hedge. Theoretically, if I’m trading one set of cash flows for another set of cash flows that is equal in value, and the risk of each side of the swap is balanced, then the cost of my swap should approach zero (although I will still need to deposit capital to cover the risk of portfolio fluctuations).
One way to make the portfolio look balanced and low risk is to make it (1) diversified and (2) “market neutral.” That is, I want lots of stocks in my portfolio so they won’t all be moving in the same direction at the same time (or so my dealer thinks). Furthermore, I’m going to do something else - I’m going to make the long and short portfolios equal in size. That way, if the market as a whole drags down all the stocks, whatever I lose on the long side should be made up by what i gain on the short side (and vice versa). It turns out, using the list of big mover stocks, I need about $125 billion in notional long exposure and $125 billion in matching notional short exposure, starting on 12/31/2020, weighted in all the stocks by their starting market cap, to end up with my predicted $100 billion gain on the transaction.
Remember when I told you this wouldn’t work? Here’s why: I am effectively shorting about 8% of market cap of the short portfolio. That’s not too bad. There is no limit to how many swap contracts can be written on a particular underlying security. During the financial crisis, positions in some swaps aggregated across all portfolios reached 10x or 20x the actual value of the underlying. 8% is a lot of interest to take in a security, but its within the range of what my dealer can hedge by bringing in other dealers and using other instruments. Unfortunately, to balance that portfolio, I need to take a long interest equal to 350% of the winning stocks, which are much smaller and less liquid that the losers. That risk will be almost impossible for my dealer to hedge at reasonable cost, so he is going to have to absorb a lot of that risk himself. He will both demand that I post a lot of margin capital and he will effectively charge me a lot of cash just to enter the trade. So, to get my notional $250 billion total return swap, he might effectively demand billions up front and many more billions in margin. More likely, he will just say that it’s too much interest to hedge but he can hedge 5% of it, so at huge cost, he will sell me a hedge that I will make $10 billion on. I don’t know what kind of measly country I can buy for $10 billion and I don’t want it, so I have to rethink.
So I want a portfolio that looks safer than the last one, that is easier for the dealer to hedge. The real answer is I want a total return swap limited to only the most liquid names: the 100 highest market capitalization stocks in the US.
You might think I will short all losers in the list and go long on all the winners, but again, that would be wrong. There more winners than losers YTD as I write this. I want my portfolio to be market neutral, so I actually have to short some stocks that turn out to make money in order to balance my long exposure to other, better stocks. My total return swap will need long exposure equal to 7% of the long portfolio and 7% short exposure to the short portfolio.
The notional value of this portfolio is $1.5 trillion, which is huge, but it is market neutral, appears low risk, and is relatively cheap to hedge. It is within the scope of a swaps portfolio that a $10 billion institution might have. I think if you had $10 billion to start with, you could probably get a dealer to sell you this swap, which in our world would have made $200 billion.
Long list of individual reference securities spoilered:
Summary
The long portfolio would be:
HON
DHR
PYPL
MSFT
ADP
MU
CSCO
EL
INTC
SPGI
IBM
TXN
MMM
INTU
PM
LLY
HD
BA
PLD
LOW
MO
FB
BLK
ANTM
ORCL
CI
NVDA
RTX
BRK-A
CVS
CVX
SNAP
UPS
TGT
C
USB
JPM
PNC
GE
UNH
CAT
MS
AXP
DE
LRCX
GOOG
SCHW
BAC
GM
GS
XOM
WFC
AMAT
The short portfolio would be:
NOW
QCOM
TSLA
AMD
MRK
NFLX
NEE
AAPL
T
NKE
PG
TMO
VZ
AMT
ZM
ABT
WMT
ABNB
AMZN
DIS
ADBE
PEP
UBER
COST
KO
CMCSA
MA
AMGN
ISRG
SQ
V
CHTR
BKNG
ABBV
In order to trade swaps, I need to be an eligible contract participant, or “ECP,” which for our purposes means that I only need a minimum of $10 million in assets.
Does it break the rules to become an angel investor? That might be a way to move from the $1-2 billion game into the richest in the world department. It would take some time, though, but potentially only a few years knowing ahead of time which startups to back, and when to diversify out of them.
If you setup a venture capital company with your money, and then gave them a few “tips” it might even look completely legitimate. There would be plenty of losers, but a few giant winners, just like a regular successful venture capital company.
What game? I thought the hypothetical involved starting off with only one dollar, and getting more dollars solely by putting money down this stock or that stock or some other stock. Sure, once he gets enough dollars to really make stuff happen he wouldn’t need to win any more money off of me; but until then, we can engage in transactions where he always wins money off of me and ‘figuring out my game’ is beside the point; he could know exactly what I’m doing — I could tell him exactly what I’m doing — and still he could build his one dollar into many dollars instead of wasting any of his time or money ‘fucking with me by letting me win’.
Going back to my data forensics days, the pattern of the trades themselves could be used as evidence. Which is not to say would necessarily be convicted. That was one of the things we used to do. Look for correlations and improbable patterns.